Multiscale collaborative process optimization method for automated fiber placement

Considering the multiscale effect of tow structure and defect evolution under Automated Fiber Placement (AFP), a novel multiscale collaborative process optimization method is proposed in this paper. In macro-and meso-scale, finite element (FE) method is used to obtain all stored strain energy of the FE model and the strain energy of meso-unit under different process parameters. Viscosities, the diffusion coefficient of resin along the interface, and the adsorption energy of the interface are gained using the molecular dynamics (MD) method at microscale. Characteristic parameters of stress waves and defects are quantitatively identified through experiments. Furthermore, the gray correlation analysis and analytic hierarchy process are used to establish the constraint conditions of optimization. Process optimization is then performed using the response surface method. Finally, the comprehensive advantages of the optimal process parameters are evaluated. The results show that the optimal process parameters can improve the comprehensive multiscale mechanical performance significantly.